Pith. sign in

REVIEW 4 major objections 4 minor 41 references

Highly Accurate and Diverse Traffic Data: The DeepScenario Open 3D Dataset

T0 review · 4 major / 4 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read First public drone dataset offers full-3D traffic trajectories for 14 classes.

desk verdict Genuinely new 3D drone trajectory dataset, but the 4.8 cm accuracy claim is circular until validated on a held-out set. read the letter →

arxiv 2504.17371 v3 pith:N6JF2I2V submitted 2025-04-24 cs.CV

Oussema Dhaouadi , Johannes Meier , Luca Wahl , Jacques Kaiser , Luca Scalerandi , Nick Wandelburg , Zhuolun Zhou , Nijanthan Berinpanathan
show 2 more authors
Holger Banzhaf Daniel Cremers
This is my paper · ORCID
classification cs.CV
keywords 3Dtrajectorydatasetdronedatacollectionautonomousdriving6DoFboundingboxesmonoculardetectiontrafficparticipantsscenariominingmotionprediction
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper introduces DSC3D, a drone-captured dataset of traffic movement with complete 3D bounding-box annotations for 14 categories across five locations in Germany and the United States. It is presented as the first publicly available dataset of its kind, with more than 175,000 trajectories, about 15 hours of footage, and a claimed median positional error of 4.8 centimeters. The need arises because existing drone datasets provide only 2D or 2.5D tracks, while vehicle-mounted sensors suffer from occlusion and limited range; 3D trajectories would let planners, predictors, and simulators reason about true spatial relations. The paper demonstrates that adding scenarios extracted from the dataset to an established training set improves motion-prediction error by roughly 44%, and that the data support scenario mining and generative traffic agents.

What carries the argument

The load-bearing mechanism is a monocular ground-aware 3D tracking pipeline: the scene is first reconstructed with structure-from-motion and multi-view stereo into a geo-referenced 3D mesh, and a NURBS ground surface is fitted to the road; each frame's camera pose is obtained by feature-matching the frame to the reconstructed scene and solving PnP within RANSAC, then smoothed with a Kalman filter. Monocular detections (2D box, class, dimensions, orientation, depth) are refined by intersecting the camera ray through the detected ground center with the ground mesh, and by re-aligning the object's orientation to the ground normal. This converts per-frame pixel detections into metric 3D boxes that remain consistent across frames, after a Kalman filter and RTS smoother produce the final trajectories.

What would settle it

Re-capture a subset of the five scenes with surveyed ground-control points and high-precision references such as RTK-GPS on instrumented vehicles, then compare the released trajectories against those independent positions; if the median error substantially exceeds 4.8 cm, the accuracy claim collapses.

Watch

Extended reading notes

Core claim

DSC3D is the first publicly available drone-based trajectory dataset in which every tracked object carries a complete 6-degree-of-freedom 3D bounding box with metric dimensions, in geo-referenced coordinates, for 14 classes. The dataset covers a parking lot, a pedestrian-dense inner-city street, an unsignalized steep intersection, a federal highway, and a suburban intersection, yielding 177,151 unique trajectories over about 15 hours. The paper claims that its monocular tracking pipeline, which refines single-camera detections against a reconstructed ground mesh, achieves a median positional error of 4.8 cm when compared with manual annotations, and a 97% NuScenes-style score evaluated in 3D space. It further argues that this level of accuracy and 3D completeness is what makes the data useful for motion prediction, planning, scenario mining, and reactive traffic simulation.

Load-bearing premise

The reported centimeter-level accuracy assumes the monocular detector and tracker produce trajectories reliable enough to serve as ground truth, even though the detector was trained on images from this same dataset and the error is measured against manual annotations of those same recordings.

Editorial extensions

If this is right

  • Motion-prediction and planning models can be trained directly on metric 3D trajectories with true object sizes and orientations, enabling reasoning on non-planar surfaces like slopes and multi-level interactions.
  • The dataset's 14-class taxonomy and its long trajectories (up to 984 s) support scenario mining for safety-critical events, including parking maneuvers and pedestrian-heavy urban interactions.
  • Adding extracted scenarios from DSC3D to an existing driving dataset improves average displacement error by 44.1% and final displacement error by 44.3%, indicating that drone 3D data can improve generalization of current predictors.
  • The geo-referenced HD maps and 3D mesh make the dataset directly usable for simulation, digital twins, and closed-loop generative traffic agents.
  • Occlusion-free drone perspectives allow nearly every vehicle in a scene to serve as a potential ego vehicle, expanding the volume of usable training scenarios per hour of recording.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The 4.8 cm median error is an internal estimate: it is computed against manual annotations that the detector itself was trained on, so users should treat it as a claim to verify independently rather than an established ground-truth accuracy.
  • If the accuracy claim survives independent checks, the pipeline's recipe—monocular detection plus reconstructed ground mesh—could be applied to convert existing drone video archives into 3D trajectory data at scale, not just for the five locations shown.
  • The inclusion of slopes up to 20% offers a natural testbed for whether current flat-ground prediction models fail on non-planar scenes, a question the paper does not directly pursue.
  • Public release with an interactive visualizer invites community scrutiny of annotation quality, which could either confirm or undermine the claimed accuracy more effectively than the paper's own validation.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper introduces DSC3D, a drone-captured traffic dataset with 15 hours of footage, 177,151 unique trajectories, 14 object categories, and full 3D annotations (6DoF boxes, dimensions, orientations) across five locations in Germany and the United States. The authors describe a monocular pipeline involving geo-referenced 3D scene reconstruction, ground mesh generation, camera calibration, GroundMix-based 3D detection, detection refinement, and Kalman-filter tracking. They claim a median positional accuracy of 4.8 cm, and they demonstrate applications in motion prediction, scenario mining, and generative traffic agents. The dataset and an interactive visualization platform are publicly available.

Significance. If the correctness of the annotations can be established, DSC3D would be a valuable community resource: it is the first drone-based dataset to offer complete 3D trajectories with 6DoF poses and metric dimensions, and its scale and class diversity substantially exceed prior drone trajectory datasets. The public release, the interactive visualization interface, and the downstream evaluations on prediction and scenario mining are strengths. However, the headline accuracy claim of 4.8 cm is not currently supported by an independent validation protocol, and because accuracy is a central advertised property of the dataset, this issue is load-bearing for the paper's main contribution.

major comments (4)
  1. [Section IV-B] The accuracy validation is circular as described. Section IV-B states that the positional median error is computed by comparing predicted detections with manually annotated detections, and it explicitly notes that 'images of this dataset were included in the training of the detection network.' Section III-F adds that the detection model is finetuned on 'highly curated, manually-annotated images' from the same data. Without a track-disjoint or scene-disjoint split, and without an explicit statement that the validation annotations were never used in training or finetuning, the 4.8 cm figure measures the model's fit to its own training distribution rather than the absolute accuracy of the derived trajectories. Please provide the evaluation protocol in full: how the manual annotations were partitioned, how many objects were annotated, and how the 4.8 cm median was computed from the per-frame or per-trajectory errors.
  2. [Section III-B, III-D, IV-B] The manual annotations are not an independent ground truth. Both the predicted detections and the reference annotations are derived from the same monocular images and the same reconstructed ground mesh, so systematic errors in the mesh (15 cm 3D RMSE) and the camera poses (1.9 m GPS RMSE) propagate into both sides of the comparison. A self-consistent comparison can therefore report small relative errors while the absolute positions are significantly less accurate. To support the 4.8 cm claim, the authors should compare against an external absolute reference (for example RTK-GPS on a survey vehicle, LiDAR scans, or surveyed ground control points) or at least quantify the worst-case error contribution of the 15 cm mesh error and 1.9 m pose error to object positions.
  3. [Abstract, Section IV-B, Table II] The paper's headline number is not reproducible from the reported experiments. The abstract and contributions state a median positional error of 4.8 cm, but Section IV-B only reports a 'NuScenes score of 97%' and does not define the positional error metric, the number of validation frames or objects, or a per-location breakdown. Table II reports RMSE values for GPS and reconstruction but not the 4.8 cm detection error. Please add a dedicated accuracy subsection with a precise definition of the positional error (e.g., 3D center error versus ground-center error), the evaluation split, per-location results, and the number of annotated tracks used for validation.
  4. [Section III-F] The reference for the RTS-Smoother is incorrect: the text cites '[7]' (the exiD paper) for an RTS-smoother, but the exiD paper is a dataset paper and does not introduce this technique. Please cite the original Rauch-Tung-Striebel smoother reference or a method paper that describes the smoothing step, and clarify how smoothing affects the reported accuracy figures.
minor comments (4)
  1. [Section IV-C, Table III] The dataset name is inconsistent: the text and Table I use 'AD4CHE' while Table III uses 'AD4ACHE'; please unify the spelling.
  2. [References] Reference [36] contains a typo: 'by aearial imagery' should be 'by aerial imagery'. Please also verify the authors and venue details of this reference.
  3. [Section IV-B] The term 'NuScenes score of 97%' is ambiguous; the nuScenes detection benchmark uses a specific metric (NDS), and it is unclear how that metric is adapted to this monocular drone setting. Please define the score and its range.
  4. [Tables and Figures] In Table I, the symbol '?' for CitySim's video availability is undefined in the caption; please either define it or replace it with a clear yes/no/partial indicator.

Circularity Check

2 steps flagged · score 6.0 of 10

The headline 4.8 cm accuracy claim is an in-sample fit statistic: the detector is finetuned on the same dataset whose manual annotations are used as the validation reference, and the reconstruction 'error' is the minimized BA residual.

  1. fitted input called prediction [Section IV-B (Accuracy); Section III-F (3D Object Tracking)]
    ""The detection validation was performed by comparing the positional median error between predicted detections and manually annotated detections of the same objects. ... Note that images of this dataset were included in the training of the detection network." "Further, we finetune our 3D detection model with highly curated, manually-annotated images to make sure the quality of the data is high.""

    The paper's headline accuracy claim (median positional error of 4.8 cm, listed as a main contribution) is presented as a validation of the tracking pipeline. But the predicted detections come from a network finetuned on manually annotated images of this very dataset, and the reference is manual annotations of the same objects. With no held-out, track-disjoint, or location-disjoint split reported, the comparison measures the model's fit to its own training labels rather than trajectory accuracy against independent ground truth. The 4.8 cm figure therefore reduces by construction to a training-set residual, not an external accuracy estimate.

  2. fitted input called prediction [Section IV-B (Accuracy)]
    ""The Root Mean Squared Error (RMSE) values, derived from BA termination residuals, show a 1.928 meters deviation between reconstructed poses and consumer-grade GPS coordinates, and a 3D reconstruction error of 15 centimeters.""

    The reported '3D reconstruction error' and 'GPS RMSE' are derived from bundle-adjustment termination residuals. The BA objective in Eq. (1) is L = L_reproj + λ Σ ||c_Ii - g_Ii_local||^2, i.e., the exact quantities being reported are the ones BA minimizes. Reporting the minimized objective value as an accuracy estimate against independent ground truth is circular: the number is a fit residual by construction, not a measurement relative to an external reference, and systematic mesh or pose errors remain invisible to this self-consistency check.

full rationale

The circularity is concentrated in the accuracy validation, which is a central contribution of the paper. The 4.8 cm median positional error is computed by comparing detector outputs to manual annotations of the same objects while the detector was finetuned on images of the same dataset; no independent held-out set is described. Likewise, the reconstruction accuracy is reported from BA termination residuals, i.e., from the very objective that was minimized, so the '15 cm 3D reconstruction error' is not an independent measurement. These two steps reduce the paper's quantitative accuracy claims to fit statistics. By contrast, the dataset's scale, class diversity, geographic spread, and downstream application demonstrations are externally checkable and not circular; the use of the authors' prior GroundMix and FlexRoad components is legitimate building-block reuse rather than a self-citation chain bearing the central claim. Because the central quantitative claims are undermined by in-sample validation, a score of 6 is appropriate rather than a lower score.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The central claim (a large 3D trajectory dataset) relies on standard reconstruction and detection methods. There are no invented physical entities. The main unproven inputs are the accuracy of the ground mesh and the GPS alignment, and the validity of the in-dataset accuracy evaluation.

free parameters (2)
  • lambda (GPS alignment weight in joint BA)
    Eq. (1) balances reprojection error and GPS constraint; the value is not reported, affecting geo-referencing accuracy.
  • Detection and tracking hyperparameters
    Kalman filter noise, RTS smoothing, and detection thresholds are not specified; trajectory smoothness and accuracy depend on these.
assumptions (3)
  • domain assumption The reconstructed ground NURBS mesh accurately represents the road surface for all locations.
    Used in Section III-E2 to refine detected 3D positions by ray-ground intersection; errors propagate to all 3D boxes.
  • domain assumption Consumer-grade GPS provides sufficient geo-referencing accuracy.
    Section III-B and IV-B report GPS RMSE of 1.928 m, which may limit absolute positioning accuracy despite 4.8 cm relative accuracy claims.
  • domain assumption The monocular 3D detection network (GroundMix) produces unbiased 3D boxes on the target scenes.
    Section III-E relies on GroundMix; the model is finetuned on manually annotated images from this dataset, and no independent test set is used.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Highly Accurate and Diverse Traffic Data: The DeepScenario Open 3D Dataset." pith.science (2026). https://pith.science/paper/N6JF2I2V

@misc{pith2026250417371,
  author       = {Pith},
  title        = {Pith review of: Highly Accurate and Diverse Traffic Data: The DeepScenario Open 3D Dataset},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/N6JF2I2V}},
  note         = {Machine review of arXiv:2504.17371}
}
read the original abstract

Accurate 3D trajectory data is crucial for advancing autonomous driving. Yet, traditional datasets are usually captured by fixed sensors mounted on a car and are susceptible to occlusion. Additionally, such an approach can precisely reconstruct the dynamic environment in the close vicinity of the measurement vehicle only, while neglecting objects that are further away. In this paper, we introduce the DeepScenario Open 3D Dataset (DSC3D), a high-quality, occlusion-free dataset of 6 degrees of freedom bounding box trajectories acquired through a novel monocular camera drone tracking pipeline. Our dataset includes more than 175,000 trajectories of 14 types of traffic participants and significantly exceeds existing datasets in terms of diversity and scale, containing many unprecedented scenarios such as complex vehicle-pedestrian interaction on highly populated urban streets and comprehensive parking maneuvers from entry to exit. DSC3D dataset was captured in five various locations in Europe and the United States and include: a parking lot, a crowded inner-city, a steep urban intersection, a federal highway, and a suburban intersection. Our 3D trajectory dataset aims to enhance autonomous driving systems by providing detailed environmental 3D representations, which could lead to improved obstacle interactions and safety. We demonstrate its utility across multiple applications including motion prediction, motion planning, scenario mining, and generative reactive traffic agents. Our interactive online visualization platform and the complete dataset are publicly available at https://app.deepscenario.com, facilitating research in motion prediction, behavior modeling, and safety validation.

Figures

Figures reproduced from arXiv: 2504.17371 by the authors.

Figure 1
Figure 1. Visualization of traffic participants across five locations in our DeepScenario Open 3D Dataset ( [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Our pipeline includes (A) data collection, (B) geo-referenced 3D scene reconstruction and ground generation, (C) map creation, (D) recording [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Average of trajectory duration per class across datasets. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Average trajectory distances per class across datasets. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 6
Figure 6. Figure 6: Distribution of Time-to-Park (left) and Number of Direction [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 9
Figure 9. Figure 9: Synthetic scenarios generated by traffic agents on [PITH_FULL_IMAGE:figures/full_fig_p007_9.png]
Figure 8
Figure 8. Figure 8: exemplifies these results by showing representa￾tive examples for both a critical scenario at STR with low PET and a typical parking maneuver with direction switches at SIFI. These distributions and examples therefore suggest that our dataset covers both routine parkin…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 34 canonical work pages

  1. [7]

    The exid dataset: A real-world trajectory dataset of highly interactive highway scenarios in germany,

    T. Moers, L. Vater, R. Krajewski, J. Bock, A. Zlocki, and L. Eckstein, “The exid dataset: A real-world trajectory dataset of highly interactive highway scenarios in germany,” in IV, 2022, pp. 958–964

    work page 2022

  2. [1]

    The ind dataset: A drone dataset of naturalistic road user trajectories at german intersections,

    J. Bock, R. Krajewski, T. Moers, S. Runde, L. Vater, and L. Eckstein, “The ind dataset: A drone dataset of naturalistic road user trajectories at german intersections,” in IV, 2020, pp. 1929–1934

    work page 2020

  3. [2]

    The round dataset: A drone dataset of road user trajectories at roundabouts in germany,

    R. Krajewski, T. Moers, J. Bock, L. Vater, and L. Eckstein, “The round dataset: A drone dataset of road user trajectories at roundabouts in germany,” in ITSC, 2020, pp. 1–6

    work page 2020

  4. [3]

    openDD: A large-scale roundabout drone dataset,

    A. Breuer, J. Term ¨ohlen, S. Homoceanu, and T. Fingscheidt, “openDD: A large-scale roundabout drone dataset,” in ITSC, 2020, pp. 1–6

    work page 2020

  5. [4]

    Citysim: A drone-based vehicle trajectory dataset for safety-oriented research and digital twins,

    O. Zheng, M. Abdel-Aty, L. Yue, A. Abdelraouf, Z. Wang, and N. Mahmoud, “Citysim: A drone-based vehicle trajectory dataset for safety-oriented research and digital twins,” Transportation Research Record, vol. 2678, no. 4, pp. 606–621, 2024

    2024

  6. [5]

    The highd dataset: A drone dataset of naturalistic vehicle trajectories on german highways for validation of highly automated driving systems,

    R. Krajewski, J. Bock, L. Kloeker, and L. Eckstein, “The highd dataset: A drone dataset of naturalistic vehicle trajectories on german highways for validation of highly automated driving systems,” in ITSC, 2018, pp. 2118–2125

    work page 2018

  7. [6]

    Opendrive: Open dynamic road information for vehicle environment,

    OpenDrive, “Opendrive: Open dynamic road information for vehicle environment,” https://www.opendrive.com, 2000, software tool

    work page 2000

  8. [8]

    Are we ready for autonomous driving? the KITTI vision benchmark suite,

    A. Geiger, P. Lenz, and R. Urtasun, “Are we ready for autonomous driving? the KITTI vision benchmark suite,” in CVPR, 2012, pp. 3354–3361. [Online]. Available: https://doi.org/10.1109/CVPR.2012. 6248074

    doi:10.1109/cvpr.2012 2012

Show all 41 references
  1. [9]

    Scalability in perception for autonomous driving: Waymo open dataset,

    P. Sun, H. Kretzschmar, X. Dotiwalla, A. Chouard, V . Patnaik, P. Tsui, J. Guo, Y . Zhou, Y . Chai, B. Caine, V . Vasudevan, W. Han, J. Ngiam, H. Zhao, A. Timofeev, S. Ettinger, M. Krivokon, A. Gao, A. Joshi, Y . Zhang, J. Shlens, Z. Chen, and D. Anguelov, “Scalability in perc...

    2020

  2. [10]

    nuscenes: A multimodal dataset for autonomous driving,

    H. Caesar, V . Bankiti, A. H. Lang, S. V ora, V . E. Liong, Q. Xu, A. Krishnan, Y . Pan, G. Baldan, and O. Beijbom, “nuscenes: A multimodal dataset for autonomous driving,” in CVPR, 2020, pp. 11 618–11 628

    2020

  3. [11]

    One thousand and one hours: Self-driving motion prediction dataset,

    J. Houston, G. Zuidhof, L. Bergamini, Y . Ye, L. Chen, A. Jain, S. Omari, V . Iglovikov, and P. Ondruska, “One thousand and one hours: Self-driving motion prediction dataset,” in CoRL, J. Kober, F. Ramos, and C. J. Tomlin, Eds., 2020, pp. 409–418

    2020

  4. [12]

    Argoverse 2: Next generation datasets for self-driving perception and forecasting,

    B. Wilson, W. Qi, T. Agarwal, J. Lambert, J. Singh, S. Khandelwal, B. Pan, R. Kumar, A. Hartnett, J. K. Pontes, D. Ramanan, P. Carr, and J. Hays, “Argoverse 2: Next generation datasets for self-driving perception and forecasting,” in NeurIPS, 2021

    2021

  5. [13]

    Womd-lidar: Raw sensor dataset benchmark for motion forecasting,

    K. Chen, R. Ge, H. Qiu, R. Ai-Rfou, C. R. Qi, X. Zhou, Z. Yang, S. Ettinger, P. Sun, Z. Leng, M. Baniodeh, I. Bogun, W. Wang, M. Tan, and D. Anguelov, “Womd-lidar: Raw sensor dataset benchmark for motion forecasting,” in ICRA, 2024, pp. 4766–4773

    2024

  6. [14]

    Interaction of autonomous and manually controlled vehicles multi- scenario vehicle interaction dataset,

    N. Certad, E. del Re, H. Kornd ¨orfer, G. Schr¨oder, W. Morales-Alvarez, S. Tschernuth, D. Gankhuyag, L. del Re, and C. Olaverri-Monreal, “Interaction of autonomous and manually controlled vehicles multi- scenario vehicle interaction dataset,” IEEE Intelligent Transportation S...

    2024

  7. [15]

    Next Generation Simulation (NGSIM) Vehicle Trajectories and Supporting Data,

    U. S. D. of Transportation Federal Highway Administration, “Next Generation Simulation (NGSIM) Vehicle Trajectories and Supporting Data,” https://data.transportation.gov/d/8ect-6jqj, 2016, [Accessed 16- 12-2024]

    2016

  8. [16]

    You’ll never walk alone: Modeling social behavior for multi-target tracking,

    S. Pellegrini, A. Ess, K. Schindler, and L. V . Gool, “You’ll never walk alone: Modeling social behavior for multi-target tracking,” in ICCV, 2009, pp. 261–268

    2009

  9. [17]

    V2AIX: A multi- modal real-world dataset of ETSI ITS V2X messages in public road traffic,

    G. K ¨uppers, J. Busch, L. Reiher, and L. Eckstein, “V2AIX: A multi- modal real-world dataset of ETSI ITS V2X messages in public road traffic,” CoRR, 2024

    2024

  10. [18]

    V2x-seq: A large-scale sequential dataset for vehicle-infrastructure cooperative perception and forecasting,

    H. Yu, W. Yang, H. Ruan, Z. Yang, Y . Tang, X. Gao, X. Hao, Y . Shi, Y . Pan, N. Sun, J. Song, J. Yuan, P. Luo, and Z. Nie, “V2x-seq: A large-scale sequential dataset for vehicle-infrastructure cooperative perception and forecasting,” in CVPR, 2023, pp. 5486–5495

    2023

  11. [19]

    MONA: the munich motion dataset of natural driving,

    L. Gressenbuch, K. Esterle, T. Kessler, and M. Althoff, “MONA: the munich motion dataset of natural driving,” in ITSC, 2022, pp. 2093– 2100

    2022

  12. [20]

    Learning social etiquette: Human trajectory understanding in crowded scenes,

    A. Robicquet, A. Sadeghian, A. Alahi, and S. Savarese, “Learning social etiquette: Human trajectory understanding in crowded scenes,” in ECCV, vol. 9912, 2016, pp. 549–565

    2016

  13. [21]

    Ctv-dataset: A shared space drone dataset for cyclist-road user interaction derived from campus experiments,

    A. Mukbil, Y . M. Yousif, S. Hossain, and J. P. M ¨uller, “Ctv-dataset: A shared space drone dataset for cyclist-road user interaction derived from campus experiments,” in ITSC, 2023, pp. 3186–3191

    2023

  14. [22]

    Top-view trajectories: A pedestrian dataset of vehicle-crowd interaction from controlled experiments and crowded campus,

    D. Yang, L. Li, K. A. Redmill, and ¨U. ¨Ozg¨uner, “Top-view trajectories: A pedestrian dataset of vehicle-crowd interaction from controlled experiments and crowded campus,” in IV, 2019, pp. 899–904

    2019

  15. [23]

    The JKU DORA traffic dataset,

    P. Tkachenko, N. Certad, G. Singer, C. Olaverri-Monreal, and L. del Re, “The JKU DORA traffic dataset,” IEEE Access , vol. 10, pp. 92 673–92 680, 2022

    2022

  16. [24]

    AUTOMATUM DATA: drone- based highway dataset for the development and validation of auto- mated driving software for research and commercial applications,

    P. Spannaus, P. Zechel, and K. Lenz, “AUTOMATUM DATA: drone- based highway dataset for the development and validation of auto- mated driving software for research and commercial applications,” in IV, 2021, pp. 1372–1377

    2021

  17. [25]

    Vehicle trajectory dataset from drone videos including off-ramp and congested traffic – analysis of data quality, traffic flow, and accident risk,

    M. Berghaus, S. Lamberty, J. Ehlers, E. Kall ´o, and M. Oeser, “Vehicle trajectory dataset from drone videos including off-ramp and congested traffic – analysis of data quality, traffic flow, and accident risk,” Communications in Transportation Research, vol. 4, p. 100133, 2024

    2024

  18. [26]

    The AD4CHE dataset and its application in typical congestion scenarios of traffic jam pilot systems,

    Y . Zhang, C. Wang, R. Yu, L. Wang, W. Quan, Y . Gao, and P. Li, “The AD4CHE dataset and its application in typical congestion scenarios of traffic jam pilot systems,” IEEE Trans. Intell. Veh. , vol. 8, no. 5, pp. 3312–3323, 2023

    2023

  19. [27]

    SIND: A drone dataset at signalized intersection in china,

    Y . Xu, W. Shao, J. Li, K. Yang, W. Wang, H. Huang, C. Lv, and H. Wang, “SIND: A drone dataset at signalized intersection in china,” in ITSC, 2022, pp. 2471–2478

    2022

  20. [28]

    INTERACTION dataset: An international, adversarial and cooperative motion dataset in interactive driving scenarios with semantic maps,

    W. Zhan, L. Sun, D. Wang, H. Shi, A. Clausse, M. Naumann, J. K ¨ummerle, H. K ¨onigshof, C. Stiller, A. de La Fortelle, and M. Tomizuka, “INTERACTION dataset: An international, adversarial and cooperative motion dataset in interactive driving scenarios with semantic maps,” CoRR, 2019

    2019

  21. [29]

    Shape your ground: Refining road surfaces beyond planar representations,

    O. Dhaouadi, J. Meier, J. Kaiser, and D. Cremers, “Shape your ground: Refining road surfaces beyond planar representations,” in 2025 IEEE Intelligent Vehicles Symposium. IEEE, 2025

    2025

  22. [30]

    Piegl and W

    L. Piegl and W. Tiller, The NURBS Book (2nd ed.), Berlin, Heidelberg, 1997

    1997

  23. [31]

    A visual slam-based approach for calibration of distributed camera networks,

    T. Pollok and E. Monari, “A visual slam-based approach for calibration of distributed camera networks,” in 2016 13th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS). IEEE, 2016, pp. 429–437

    2016

  24. [32]

    Loftr: Detector- free local feature matching with transformers,

    J. Sun, Z. Shen, Y . Wang, H. Bao, and X. Zhou, “Loftr: Detector- free local feature matching with transformers,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2021, pp. 8922–8931

    2021

  25. [33]

    Lightglue: Local feature matching at light speed,

    P. Lindenberger, P.-E. Sarlin, and M. Pollefeys, “Lightglue: Local feature matching at light speed,” in ICCV, 2023, pp. 17 627–17 638

    2023

  26. [34]

    Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography,

    M. A. Fischler and R. C. Bolles, “Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography,” Communications of the ACM , vol. 24, no. 6, pp. 381–395, 1981

    1981

  27. [35]

    CARLA Drone: monocular 3d object detection from a different perspective,

    J. Meier, L. Scalerandi, O. Dhaouadi, J. Kaiser, A. Nikita, and D. Cremers, “CARLA Drone: monocular 3d object detection from a different perspective,” in GCPR, 2024

    2024

  28. [36]

    DeepUrban: Interaction-aware trajectory prediction and planning for automated driving by aearial imagery,

    C. Selzer and F. Flohr, “DeepUrban: Interaction-aware trajectory prediction and planning for automated driving by aearial imagery,” in ITSC, 2024

    2024

  29. [37]

    Traffic and safety rule compliance of humans in diverse driving situations,

    M. Kurenkov, S. Marvi1, J. Schmidt, C. B. Rist1, A. Canevaro1, H. Yu1, J. Jordan, G. Schildbach, and A. Valada, “Traffic and safety rule compliance of humans in diverse driving situations,” in CoRL workshop, 2024

    2024

  30. [38]

    Behaviorgpt: Smart agent simulation for autonomous driving with next-patch prediction,

    Z. Zhou, H. Hu, X. Chen, J. Wang, N. Guan, K. Wu, Y .-H. Li, Y .-K. Huang, and C. J. Xue, “Behaviorgpt: Smart agent simulation for autonomous driving with next-patch prediction,” 2024. [Online]. Available: https://arxiv.org/abs/2405.17372

    2024 arXiv

  31. [39]

    Versatile behavior diffusion for generalized traffic agent simulation,

    Z. Huang, Z. Zhang, A. Vaidya, Y . Chen, C. Lv, and J. F. Fisac, “Versatile behavior diffusion for generalized traffic agent simulation,”

  32. [40]

    Trafficbots v1.5: Traffic simulation via conditional vaes and transformers with relative pose encoding,

    Z. Zhang, C. Sakaridis, and L. V . Gool, “Trafficbots v1.5: Traffic simulation via conditional vaes and transformers with relative pose encoding,” 2024. [Online]. Available: https://arxiv.org/abs/2406.10898

    2024 arXiv

  33. [2024]

    Available: https://arxiv.org/abs/2404.02524

    [Online]. Available: https://arxiv.org/abs/2404.02524

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.